Economic Information and Law
Paper for the New Zealand Association of Economists Conference
Christchurch, 27-29 June 2001
By Stuart Birks and Gary Buurman[1]
1. Introduction
Many aspects of
law relate to economic policy considerations. Often, economic
policy interventions require law not only for enforcement, but
for interpretation. This paper considers some aspects of economic
input into economic policy debate as it relates to application of
the law. We begin with a general discussion where we take as the
starting point the State Services Commission’s (SSC’s)
project on the quality of policy advice and the implications of
positivism and the ‘scientific’ advance of economics.
We then focus on three, more specific, issues. Firstly, whether
economic evaluation (under the guise of cost-benefit analysis)
can be subject to political influence. Secondly, whether official
data (like the census) may inappropriately shape perceptions.
Thirdly, whether theory is applicable to practical problems. We
conclude by relating this discussion specifically to law and
economics.
2. The SSC and the Quality of Policy Advice
In 1997, under
the instigation of Jenny Shipley as Minister of State Services,
the SSC began a project concerned with the quality of
policy advice being tendered to Government Ministers. Three early
papers that contribute to the project are released on the web at http://www.ssc.govt.nz/documents/Occ_Papers_Contents_Screen.htm.
The project has come up with a number of issues that impact on
the quality of policy advice. Elsewhere (see Birks and Buurman,
2000) we have argued that other sectors in New Zealand are not
immune from these issues which result in misinformation in the
policy-making arena. One problem acknowledged by the SSC is that
there is little critical appraisal in New Zealand. (Occasional
Paper, No. 9, p. 5).
Some of the
general concerns of the project are:
·
Inadequate human resources in some policy units
·
Inattention to implementation issues
·
Counter-productive departmental patch protection
The project
identifies five contributing factors;
·
Lack of clarity from ministers on desired outcomes
·
Insufficient incentives for cooperation on policies that
‘cut across’ departments
·
Variations in standards of leadership
·
Under-investment in capability development leading to shortages
·
Inadequate and/or ineffective use of information, research,
evaluation and consultation techniques as policy inputs
Clearly there are
important issues here. There is a communication problem conveying
to policy advisors exactly what it is that the government’s
policies are supposed to achieve. There is also competition (seen
here as unhealthy rivalry) among public service departments.
Departments also face resourcing and leadership issues. However
economists are specifically interested in the use of research,
information and evaluation, as well as implementation issues,
although the latter are sometimes assumed away.
There are many
aspects to policy advice, including a knowledge of what works and
what doesn’t, what needs to be changed and what alternatives
are available (No. 9, p. 8). Wellington based advisors may not
have access to all these perspectives and can even be isolated
from staff in their own department. Paper number 9 (p. 8) argues
that, “As a result, advice tends to be based on theoretical
frameworks, extant information and current practice, and the
advice of ‘dependable’ sectoral interest groups”.
If advisors are unable to critically assess information from
their lobby groups, then they are vulnerable to capture. This
means that a policy that is instituted can be out of touch with
actual conditions in New Zealand.
This paper focuses on the last of the
five contributing factors above; the use of information,
evaluation and consultation techniques. Our aim here is a
discussion of some of the factors that lead to bad policy
decisions. Initially, we focus on what policy question to select
for research.
In 1953, Milton
Friedman wrote an article titled “The Methodology of
Positive Economics”, where he cited Keynes’ distinction
between positive and normative economics (p. 3). This paper acted
as a catalyst for those economists who felt that economics is
best regarded as a value free discipline which is, at least in
part, a science. According to Alvey (p. 3), most economists now
consider their discipline to be a technical one that is free from
ethical concerns and that one of the great advances of economics
was its ‘escape’ from the moral sciences. In this
‘scientific’ view, economists concentrate on positive
economic questions and the implication is that there is a large
area of agreement among practitioners on positive economic
material. The test of whether to accept or reject a theory is its
predictive power. In this context, the accuracy of assumptions is
secondary. That is, theories are not to be accepted or rejected
on the grounds of the realism (or not) of their assumptions.
In this approach,
it follows that a main role of economists in policy matters is to
undertake research and turn any results over to those responsible
for making policy decisions. It is assumed that the policy goal
is to maximize social welfare and policy makers would then decide
a course of action (where not to take action is still regarded as
a policy decision). Recall that policy interventions often
require recourse to the law, if only for interpretation. In the
context of our paper, positivism is compatible with an economist
undertaking an economic evaluation through say cost-benefit
analysis (CBA), and then turning the results over to a government
department. However there is a dilemma here in that, if the CBA
method is comprehensive, the decision is defined by the
conclusion of the analysis. Alternatively, if the result is not
definitive, leaving the final decision to policy-makers, this
suggests that something has been left out of the analysis. If
this is so, what does the CBA result mean?
Myrdal and others
have taken a different perspective than that of the positivists.
They hold that scientific exertions, particularly in economics,
are conditioned by the society in which we live. This has always
been the case and the works of Smith and Ricardo, among other
famous names, were all in response to changing political
conditions (see Myrdal, p. 5). He contends (p. 6) that while
“selective conditioning” of the choice of research
problems does not invalidate the research done, it should make us
aware of other influences on the content of the research. We
could possibly go further here. For example when researching the
‘gender pay gap’, it might be useful to ask why was a
gender grouping selected? Shouldn’t we also consider sub-categories
according to age, experience, etc.? Similarly, Chapple has
criticised the use of ethnicity as a basis for the “Closing
the Gaps” policy (Chapple, 2000).
Any research for
policy purposes must consider the policy objectives and the
available policy instruments of the particular policy-making body
concerned. More specifically, it should incorporate a description
of the causal relationship between the instruments and the
objectives. This research will therefore be institution-specific
rather than a comprehensive analysis. Looked at in this way, it
is clear why there could be many angles on the same issue.
The process is
even more constrained, as Myrdal describes (p.12-13). He argues
that research must start from theory and it is then tempting to
use familiar tools or models. But when the assumptions do not fit
the situation, there can be serious problems and the risk of
fundamental error in research conclusions can be great. A danger
with the positivist approach is that they may too readily
transfer models from one environment to another without due
regard for the significance of assumptions. Myrdal and
others would also argue that preconceptions and subjective
valuations are present in economic research. An example concerns
Maori land. Even though there had been some recognition of
different forms of land tenure by Maori, financial and legal
structures were not adjusted to accommodate this.
If we take Myrdal’s
and the positivists’ positions as extremes and consider them
from the perspective of economics, it is likely that the views of
the profession are arranged along the spectrum. It is also likely
that we would have a bimodal distribution rather than a bell-shaped
curve. For example, those leaning towards the positivist view
might emphasize positive aspects of economic theory and the
testing of theories on their predictive power. They may also
acknowledge the part played by preconceptions and valuations, but
accept this in a ‘matter of fact’ manner. Economists
sympathetic to Myrdal’s view might place much greater weight
on making their value judgements explicit at the outset of a
study as well as checking the realism of the assumptions used.
This group might also acknowledge the importance of a theory’s
predictive power and find the existence of a body of positive
knowledge useful.
A recent example highlighting the problems of transferring policy from one country to another was illustrated in the keynote address at the Law and Economics Association of New Zealand’s annual general meeting. Professor Lueck had observed sharecropping situations in the United States where outcomes were highly variable. He attempted to give a theoretical explanation for this. By his reasoning and observations, he suggested that such contracting arrangements would not arise in areas such as dairy farming. This contrasts with the New Zealand experience where sharemilking is widespread but sharecropping is not observed.
If we accept that Myrdal’s group has a valid point, then valuations will enter into the choice of approach as well as the selection of problems. An example of the choice of approach concerns the discount rate. The authors recall serious discussion at a conference some years back on whether to use a zero discount rate. The reasoning was that environmental costs and benefits of an action are likely to be observed later rather than sooner. Discounting reduces the assessed value of these costs and benefits, thus resulting in less concern for environmental factors. A zero discount rate would eliminate this problem. Perhaps a more justifiable response would be to reconsider the values used for these future costs and benefits, rather than to avoid discounting. An example of problem selection would involve which environmental problem to subject to economic evaluation. The problem selection issue could have serious ramifications for actual policy decisions where economic activity is subjected to legal regulation.
An example illustrating the significance of choice of approach can be found in a report produced jointly by Child, Youth and Family and the Ministry of Social Policy under the auspices of the Children, Young Persons, and the Families Act Research and Evaluation Fund Management Committee (Saville-Smith, 2000). The report speaks of “substantial differences in the quantity of research undertaken by researchers in relation to the key dimensions of individual, family, neighbourhood, and social system” (p.9). It noted far more research into some factors than into the others. Discussing this, readers are cautioned as follows:
“Because
certain factors and/or certain dimensions have been extensively
researched does not mean that those factors or dimensions are
more important in determining child abuse or child neglect than
those that have not been extensively researched.” (pp.9-10)
As an aside, it is also pointed out that:
“Only a tiny number, if any, of
the research reviewed could be confirmed as scientifically sound
methodologically … Nevertheless…this body of research
is, despite all its limitations, already informing policy and
service perceptions and decisions.”
(p.12)
Janet Grieve of the NZ Association of Scientists stated recently that, “Priorities for research in some areas can and are being driven by politicians or their advisers” (Grieve, 2001).
The
significance of this issue is discussed in the next section.
4. Economic Evaluation and Political
Influence
Positive and
normative issues as discussed in the previous section can be
shown to have a bearing on cost-benefit analyses. We can reason
as follows.
CBA decisions
based on the positive NPV criterion are not acceptable if there
are mutually exclusive options. More than one option may yield a
positive NPV, but only one can be undertaken. Optimal solutions
can only be found if all such options are considered and compared.
Mutually exclusive options are commonly considered to arise in
situations where there is a unique resource with alternative uses.
However the problem applies generally whenever there is a scarce
resource the price of which does not reflect that scarcity. In
other words, resource allocation is not based entirely on price.
There is some non-price rationing mechanism operating. This
raises an interesting issue.
Scarcity is
measured through comparison of supply and demand. Demand is based
on available options. Cost-benefit analyses frequently relate to
the supply of goods or services for which demand is expressed
collectively through some centralized mechanism, as with
government-funded projects. This serves as a rationing process by
limiting the options considered.
If asked, people
may be willing to support a project, but they may be reluctant to
support 100 projects. The outcome therefore depends on the
options considered. Alternatively, if a range of options is
considered with different people being asked about each option,
responses could be incorrectly based on the ceteris paribus
assumption that the only project under investigation is the one
about which a person is being asked. This differs from
observations of consumer choice, where the same range of options
can be available for everyone.
So problems can
arise if preferences are sought in situations where some options
are concealed. We should therefore consider what determines the
range of options presented. There are issues both in the
operation of markets and in public sector decisions. Why are
certain products offered and others not? How are options selected
for consideration?
Options in a
market, such as the range of goods and services on offer,
commonly arise as a result of innovation and diffusion. They are
influenced by such things as incentives for research and
development, patent legislation, tax structures, availability of
entrepreneurs and their access to capital, and the openness of
consumers to new choices. These are not considerations to be
overlooked, but many of them are some distance removed from
individual evaluations and so could possibly be disregarded in
those evaluations. We should not generalize too much to say that
the availability of options need not concern us when considering
cost-benefit studies for private sector decisions. We could be
concerned, for example, with some production decisions, such as
the location of a new plant, and possibly the availability of
goods through trade, which is influenced by government policy.
Innovation can also be constrained or guided by prevailing
paradigms.
Options in the
public sector are determined far more by the political and
bureaucratic process. In these situations it may be inappropriate
to consider that a cost-benefit analysis is objective. The choice
of options for consideration can be an important part of the
overall policy-selection process, and the results of a CBA may be
determined by the initial selection of options. Option selection
therefore merits close attention. As has been suggested above,
options may be determined by the weight of research. This in turn
may result from political decisions, and, as suggested above,
volume rather than quality may be the crucial factor for
determining impact.
5. Perceptions
and data
The way we gather data affects our view of society, our definition of problems, the policies we may consider, and the assumptions we make when assessing a situation. To illustrate the problem, consider the definition of “family” according to the New Zealand Standard Classification of Households and Families (NZSCHF)[2]. It is the classification used in the census and elsewhere and is based on households and relationships between household members. This perspective has implications where relationships are not recognized, such as for same sex couples in the past, where families are not confined to one household, such as with Maori and other extended families, and where parents of dependent children live apart. A household focus results in terms such as “two parent family” and “one parent family”, with only household members being considered to be family members. Children whose parents live apart and who spend time with both parents would be more accurately termed “two home children”. Such an option is explicitly excluded from the NZSCHF. Even when the children spend equal amounts of time at each residence, they are assumed to live in one, and the other parent is not recognized as a parent. In fact, if the parent in a child’s assigned household has repartnered, than the child is considered to be in a two parent household.
It can be quite misleading to use households as the basic unit for analysis. There are two major distorting factors. People can be active members of more than one household, and there can be sharing of resources and income between households.
If people are active in more than one household but are only considered to be members of one, as with shared parenting, then the costs of their assigned household are overstated and those of the other household(s) understated. The use of equivalence scales would be inappropriate for comparison over different household types as these households are incorrectly specified. According to the Inland Revenue Department there are about 200,000 liable parents and 300,000 children in the child support system.[3] In other words, about one child in four is recorded as having a parent living in another household.
Income and resources that are shared between households are either ignored or incorrectly identified. We can see this with child support payments, for example. These are considered as part of gross income for recipient households, which understates their true value as they are tax-free. They are not recognized at all in the payer’s household, with the effect that the gross income available to the household income is overstated by the amount paid plus associated tax.
As it is not possible to identify which households are incorrectly specified, all household data becomes suspect. This affects studies on income distribution and poverty, and can also affect attempts to identify socio-economic determinants of health, educational attainment, crime, and so on. It sets the classification of family types and thereby affects the way in which we see our society.
6. Perceptions
and economic theory
The
world may not be as economists choose to see it. In Complexity,
M Mitchell Waldrop describes economist Brian Arthur’s
experience at a 1987 Santa Fe combined workshop of economists and
physicists:
Arrow
and Anderson had asked several of the economists to give survey
talks on the standard neoclassical theory. "We were
fascinated by this structure," says Anderson, for whom
economic theory has long been an intellectual hobby. "We
wanted to learn about it."
And indeed,
as the axioms and theorems and proofs marched across the overhead
projection screen, the physicists could only be awestruck at
their counterparts' mathematical prowess—awestruck and
appalled. They had the same objection that Arthur and many other
economists had been voicing from within the field for years.
"They were almost too good," says one young physicist,
who remembers shaking his head in disbelief. "It seemed as
though they were dazzling themselves with fancy mathematics,
until they really couldn't see the forest for the trees. So much
time was being spent on trying to absorb the mathematics that I
thought they often weren't looking at what the models were for,
and what they did, and whether the underlying assumptions were
any good. In a lot of cases, what was required was just some
common sense. Maybe if they all had lower IQs, they'd have been
making some better models."
…
[Arthur
says:] "The physicists were shocked at the assumptions the
economists were making—that the test was not a match against
reality, but whether the assumptions were the common currency of
the field. I can just see Phil Anderson, laid back with a smile
on his face, saying, 'You guys really believe that?' "
The
economists, backed into a comer, would reply, "Yeah, but
this allows us to solve these problems. If you don't make these
assumptions, then you can't do anything."
And
the physicists would come right back, "Yeah, but where does
that get you—you're solving the wrong problem if that's not
reality."
(Waldrop,
1992, pp.140-2)
7. What does this have to do with law?
Many legal interventions relate to economic matters, and many economic interventions rely on law for their implementation. This raises several questions, among them the following: Does economics have appropriate information to offer? What is the quality of economic information? Where there are alternatives available, can matters be resolved through law, and do lawyers have the appropriate skills?
We can consider
some of these directly using the example of the economic concept
of incidence of taxation.
In economics a distinction is drawn between technical and distributional incidence. This is commonly seen in relation to an indirect tax. Technically, the supplier may be required to pay the tax. In other words, the supplier is the person who writes the cheque to pay the tax bill. Distributionally, the direct burden of the tax is usually shared. The supplier may be able to pass on some of the tax through a rise in the price of the product. It is the distributional incidence that is of interest to economists because it shows who, in reality, is paying what share of the tax.
These concepts
are potentially very significant in relation to law. Should laws
be specified in “technical” terms or “distributional”
terms? For example, one of a married couple may be the technical
owner of an asset, while the other has a claim to a share of the
asset under matrimonial property legislation (superannuation is
an example). In a distributional sense, ownership is shared. How
should such an asset be considered in the case of bankruptcy of
one of the partners?
In practice, we
are likely to find some laws specified in technical terms,
whereas others allow consideration of distributional dimensions.
To the extent that distributional factors are economic, we could
question whether lawyers and judges are able to adequately
address them. Even economic approaches may be only partial or
approximate. On the other hand, purely technical approaches
provide limited scope to handle complex issues. The terms “letter
of the law” and “spirit of the law” come to mind.
There are
situations where the approach is not specified in the law, but
left for interpretation through case law. Again, there may be
problems of consistency and ability to adequately interpret. For
example, it is sometimes suggested that mothers be given lighter
sentences due to the needs of their children, or that fines for
obstruction of access are not imposed because the children would
suffer. Are the same arguments used consistently in relation to
loss of fathers’ earnings due to fines or imprisonment?
As this example
illustrates, there are complex issues in relation to setting and
applying guiding principles for a society. It is important that
we have a realistic view of what can be achieved. The economic
concepts of incidence identify some of the difficulties and
limitations that we may face.
In summary, there
are limitations to economics. There are further limitations when
legal processes are used. We should therefore not be
overconfident about what we can achieve with our current
institutions. Law and economics is an important area which merits
more attention.
REFERENCES:
Alvey J (2001),
“Economics as a Moral Science”, Asymmetric
Information, Issue 10, March
Birks S and G
Buurman (2000) Research for Policy: Informing or Misleading, Centre
for Public Policy Evaluation, Massey University, Palmerston North
Chapple s (2000) Maori
socio-economic disparity
http://www.mosp.govt.nz/publications/docs/simonchapple.pdf
Friedman M (1953),
Essays in Positive Economics, University of Chicago Press,
Chicago)
Grieve J (2001)
“Stop politicians meddling in science funding”, Sunday
Star-Times, April 8, p.A10
Lueck D (2001),
“Ronald Coase goes to the Farm: Contracts and Organisation
in Agriculture” Paper presented to the Wellington AGM of the
Law and Economics Association of New Zealand
Myrdal G (1970), An
Approach to the Asian Drama: Methodological and Theoretical, New
York: Vintage
http://www.dsw.govt.nz/publications/docs/familialcaregiverslitreview.pdf
State Services
Commission (1999), Looping the Loop: Evaluating Outcomes and
Other Risky Feats, Occasional Paper 7, (State Services
Commission: Wellington)
State Services
Commission (1999), Minds Over Matter: Human Resource Issues
Affecting the Quality of Policy Advice, Occasional Paper 8, (State
Services Commission: Wellington)
State Services
Commission (1999), Essential Ingredients: Improving the
Quality of Policy Advice, Occasional Paper 9, (State Services
Commission: Wellington)
Statistics New
Zealand (1995) New Zealand Standard Classification of
Households and Families, Statistics New Zealand, Wellington
Waldrop M M (1992)
Complexity, London: Penguin
[1]
Centre for Public Policy Evaluation, Massey University,
Palmerston North
[2] Statistics New Zealand (1995)
[3] Inland Revenue figures for 31 August 1999 indicate 200,738 persons paying child support, in relation to about 300,000 children.